The long-term goal of this project is to better understand how people learn about, update, and use concepts and categories. Of central interest is the question of how goals, theories, and belief systems are combined with experience to determine conceptual behavior. Mathematical and computational models are used to pose specific questions concerning the integration of knowledge and experience and experimental observations in turn constrain theories of categorization. The proposed studies are organized around two related issues: 1. How does the organization and use of categories change with expertise? and 2. To what extent do goals and functions associated with different types of expertise lead to differing categorization schemes and different patterns of category-based reasoning? With respect to the first issue, previous work suggests that an intermediate level in a taxonomic hierarchy, known as the basic level, is psychologically salient according to a large number of converging measures. This convergence reinforces the view that categories are constrained by the correlational structure of the world and relatively impervious to both expertise in general and functional or utilitarian factors associated with expertise in particular. Our goal is to see whether and how the basic level changes with expertise. In doing so, we will also determine the influence of expertise on hierarchical levels above and below the basic level. An important component of the project is the use of multiple measures. Experiments will assess the role of types of knowledge and expertise in determining the basis of typicality, category-based induction, and other aspects of category organization. Preliminary observations suggest distinct changes with expertise in general and specific type of expertise in particular. New experiments follow-up these observations within the framework of a mathematical model (the cultural consensus model) and explore the possibility that there may be multiple privileged levels of categorization. This work is important because our current empirical generalizations and theories (particularly for biological categories) may be heavily weighted toward relatively early (novice) stages of category learning and organization. It is also important to derive a better understanding of the biological knowledge of college students. Our initial observations suggest less convergence on a single privileged level of categorization than previous research had implied. One significant aspect of this analysis is its potential to clarify how formal training in biology interacts with everyday conceptions of the biological world.